For weeks, the debate around Maju Forest and Gillman Barracks appeared to centre on a familiar tension in land-scarce Singapore: build more homes, or preserve more greenery?

On 4 August 2026, Minister of State for National Development Alvin Tan told Parliament that the current conceptual plans for both sites had already been adjusted to retain important green and heritage areas, and that conserving more would necessarily mean building fewer homes. He also emphasised that the plans for Gillman Barracks and the Sunset Way (Maju Forest) area were not final, and would continue to be refined based on public feedback. The public consultation period closed on 6 August.

That framing matters. It means the discussion is not about a fixed decision, but an evolving set of trade-offs still being worked through.

Singapore still needs new homes. If fewer can be accommodated on one site, that demand does not simply disappear.

So where do we build them instead?

And as Singapore becomes increasingly developed, could decisions such as Maju Forest and Gillman Barracks offer an early glimpse of how much harder that question will become?

Under the initial proposal for Maju Forest along Sunset Way, about two-thirds of the roughly 23-hectare forest was proposed for public housing, while the remaining areas would be retained as wildlife refuges and ecological corridors.

At Gillman Barracks, the former military site and surrounding forest have been identified for a roughly 40-hectare mixed-use residential neighbourhood comprising both public and private housing, with portions of greenery retained within the broader development.

Neither plan is final.

During the consultation and stakeholder engagements, nature groups proposed wider green corridors at both locations. At Gillman Barracks, this could improve ecological connections towards Telok Blangah Hill Park, Labrador Nature Reserve and HortPark. At Maju Forest, a more continuous corridor could strengthen habitat connectivity for wildlife. The authorities have said these suggestions will be considered as plans are refined.

It is easy to frame the situation as development on one side and conservation on the other. In reality, the planning challenge is more layered.

The Government has maintained that both sites are part of Singapore's medium-term housing pipeline. At the same time, it has noted that alternative sites raised by the public are often already earmarked for other uses, required for national needs, or not yet ready for redevelopment.

That is what makes the issue particularly interesting from a property perspective.

The question is not simply whether Singapore should build homes at Maju Forest and Gillman Barracks.

It is how many homes can reasonably be accommodated there, what should be preserved, and how the remaining housing demand is absorbed elsewhere in the system.

One thing is clear from the numbers: Singapore continues to plan for a substantial housing pipeline.

For private homes, the Confirmed List under the Government Land Sales Programme is expected to supply 9,320 units in 2026, more than 50% higher than the annual average over the previous decade.

Following the second-half 2026 land supply announcement, the overall pipeline of private residential units, including Executive Condominiums, stood at about 61,000 units, with a significant share already unsold and potentially released over the next few years.

Looking further ahead, about 60,600 private residential units including ECs are expected to be completed in the coming years, with roughly 25,900 by 2028 and another 34,700 from 2029 onwards.

The public housing pipeline is similarly substantial.

HDB plans to launch about 19,600 BTO flats in 2026 across three sales exercises, with more than 4,000 offering waiting times of under three years.

HDB is also prepared to offer more than 55,000 flats from 2025 to 2027 if necessary, depending on demand and market conditions, with 127 projects already under construction at the start of 2026.

Put together, these figures reveal an important distinction.

Singapore is not facing a shortage of housing plans.

It is facing the harder question of where a large and continuous pipeline of housing should physically go.

Singapore's housing requirement is not determined by population growth alone.

Even if the population remains stable, the number of homes required can still rise as household sizes shrink. More singles are forming independent households, young couples are setting up their own homes, and an ageing population means more seniors may live alone.

The arithmetic is simple. If 10,000 people live in households averaging four people, they require about 2,500 homes. If the same population shifts to households averaging two people, the requirement doubles to 5,000 homes.

The population has not changed. The number of households has.

This is why housing demand cannot be understood purely through population figures. Household formation, ageing, and lifestyle changes all matter.

There is also a spatial dimension.

Singaporeans tend to value proximity to transport, jobs, schools, family and amenities. This means housing supply is not just about total units, but about where those units are located.

Recent BTO projects in established estates illustrate this demand. During the parliamentary discussion, Mr Tan pointed to strong application rates for projects such as Clementi Emerald and Berlayar Rise.

The pressure is therefore not only to produce more housing.

It is to produce housing in well-connected, established locations where land is increasingly scarce.

For a small city-state, land has always been scarce. What is changing is the nature of the trade-offs.

Many future residential sites already contain existing value: forests, golf courses, industrial estates, heritage precincts, institutions, or transport infrastructure.

Singapore's planning response has increasingly focused on brownfield redevelopment: reusing already-developed land for new purposes.

The former Keppel Golf Course is a clear example. It is expected to accommodate about 10,000 public and private homes, including the Berlayar estate, where HDB flats are already entering the pipeline.

The former Old Police Academy is being transformed into the Mount Pleasant estate, which is expected to provide around 5,000 homes.

Other sites will emerge as existing uses are relocated, consolidated, or reach the end of their leases.

This approach reduces pressure on undeveloped land.

But it introduces a timing constraint.

Brownfield land does not become available on demand. Some transformations take decades. The Greater Southern Waterfront depends on port relocation to Tuas. Paya Lebar Air Base can only be redeveloped after aviation operations move in the 2030s.

This creates a structural mismatch between:

• where land will eventually become available, and
• where housing is needed in the near term

That is the underlying tension Maju Forest and Gillman Barracks sit within.

They are not just individual sites. They are part of a system where timing, not just availability, is the constraint.

Because of this timing gap, Singapore has increasingly had to make each parcel of land do more.

One approach is higher density, especially around transport nodes. This allows more homes to be built without proportionally more land.

But density only works if infrastructure scales alongside it-transport, schools, healthcare, and public spaces all need to expand in tandem.

A second approach is integration: combining housing with retail, workplaces, community spaces and transport infrastructure so that land performs multiple functions.

A third is continued redevelopment of brownfield sites as they become available.

None of these removes the need for difficult trade-offs. They simply change the framing from:

How much land do we have?

to:

How much can each site reasonably support?

The Maju and Gillman discussions are often framed as a binary choice between homes and nature. But that framing misses how people actually experience cities.

Residents do not experience housing as isolated blocks. They experience streetscapes, shade, parks, walking routes, and the sense of openness between developments.

Greenery contributes directly to that lived environment.

This becomes more important as density increases. The more people a neighbourhood accommodates, the more critical its public spaces and ecological buffers become to liveability.

Seen this way, retaining greenery is not simply a subtraction from housing supply. It is part of what makes higher-density living workable.

The challenge is balance.

Retain too little, and a neighbourhood loses character and environmental quality. Retain too much, and fewer homes can be delivered on constrained land.

There is no universal formula. Each site must be assessed on its own ecological, heritage and infrastructure context.

Gillman Barracks sits within a much larger long-term transformation of Singapore's southern coastline.

As port activities consolidate at Tuas, approximately 1,000 hectares of land and 30km of coastline will eventually be freed for redevelopment.

Source: URA

The former Keppel Golf Course is already part of this transition, with around 10,000 homes planned across public and private housing.

Source: URA, HDB

Future opportunities may also emerge around Keppel Terminal, Keppel Distripark and other port-related sites.

Gillman Barracks therefore is not an isolated parcel. It sits at the edge of a major restructuring of land use in the south.

The key question is not just how many homes it can accommodate, but how it fits into a broader district that will evolve over decades.

Its greenery, heritage buildings and arts identity are therefore not secondary considerations. They are part of what will define the character of the wider Southern Waterfront.

Maju Forest presents a different but related challenge.

Here, the issue is not heritage or industrial transition, but how development coexists with an existing natural ecosystem.

The initial plan already proposed retaining parts of the forest as ecological corridors and refuges. Public feedback has since pushed for stronger and more continuous habitat connections.

This reflects a shift in planning logic.

Instead of:

the sequence increasingly becomes:

Housing remains part of the outcome, but the natural system becomes a structuring input rather than an afterthought.

Preserving more greenery at Maju Forest or Gillman Barracks does not eliminate housing demand. It redistributes it across time and space.

Singapore's Master Plan 2025 outlines more than 80,000 public and private homes across over 10 new housing areas over the next 10 to 15 years.

New supply will come from multiple sources:

• entirely new precincts
• intensification of existing areas
• brownfield redevelopment
• and future land releases as major infrastructure shifts occur

Areas such as Dover-Medway, Newton, Paterson, Defu and the Greater Southern Waterfront are already part of this broader pipeline.

But the key point is not the list of locations.

It is that housing supply is increasingly a sequencing problem, not just a spatial one.

Some land is available now. Some will only become available later. Some are already committed to other uses.

That means every decision about one site has implications for timing elsewhere.

Most buyers will not evaluate a home based on national land-use strategy. But they are still affected by it.

A quiet field today may become a housing estate. An industrial site may become a mixed-use district. A transport node may become a dense residential hub.

This makes surrounding land use an important part of due diligence.

The URA Master Plan helps identify intended future uses. Broader transformation plans indicate where infrastructure, housing and commercial activity may expand.

But the key question is not whether change will happen.

It is what kind of change is likely, and over what timeframe.

The more useful question is therefore not:

Will this raise property values?

but:

What kind of neighbourhood is this becoming?

Maju Forest and Gillman Barracks are not isolated planning cases. They are early examples of a broader structural reality.

Singapore has a strong housing pipeline. It also has a growing number of competing land uses. And it has a long-term strategy of recycling land wherever possible.

The challenge is synchronisation.

Housing must arrive where and when it is needed. Infrastructure must scale with density. Nature and heritage must be meaningfully integrated. And some trade-offs will be unavoidable.

Part of the solution is higher density. Part is brownfield redevelopment. Part is mixed-use integration. And part is simply better sequencing of land release over time.

But the underlying tension remains.

The question is no longer just whether Singapore can build enough homes.

It is whether it can continue to do so while preserving the qualities that make those homes worth living in.

Maju Forest and Gillman Barracks are two current expressions of that challenge.

They will not be the last.